In the way we commonly understand international law lies a contradiction that cannot be concealed without complicity. It does not lie in a technical subtlety, nor in a quibble for jurists, but concerns the very substance of the idea of justice between nations. On the one hand, the inviolability of the sovereignty of states is solemnly proclaimed; on the other, the universality of human rights is affirmed. But when these two statements come into conflict, the first almost always prevails over the second. And so the question must be asked: which right is it that protects borders and abandons people?

The principle of non-interference, enshrined in the UN Charter, arose from a noble need: to prevent force from becoming an ordinary instrument of foreign policy. After two world wars, humanity wanted to erect a legal barrier against the arbitrariness of the most powerful states. The idea was simple and solemn: no one can invade another country without violating international law. And yet, in the same legal edifice, an ambiguous space was left, a grey area that today appears untenable. If a government systematically massacres part of its own population, if it exercises massive and organised violence against defenceless citizens claiming elementary rights – freedom of speech, of association, of vote – the international community, with rare exceptions, remains a spectator.

It will be said that there are instruments: sanctions, Security Council resolutions, the principle of the ‘Responsibility to Protect’ affirmed in 2005. But anyone who observes without illusions the concrete functioning of these mechanisms knows how subordinate they are to the political balances between the great powers. The veto of just one of the Security Council’s permanent members can paralyse any initiative. Thus, while external aggression is formally banned almost absolutely, internal repression can continue as long as it does not excessively upset someone’s strategic interests.

This gives rise to a disturbing distinction: killing citizens of another state is a violation of international law; killing, on a large scale, one’s own citizens is primarily an ‘internal matter’, until it becomes a threat to regional peace. It is as if the dignity of the human being is worth less than the line drawn on a map. Sovereignty, born to protect political communities, is thus transformed into a shield of the oppressor. And law, which should limit arbitrariness, ends up offering him an alibi.

Some will object that any openness to humanitarian intervention risks becoming a pretext for new forms of imperialism. It is a well-founded fear, and recent history offers controversial examples of this. But the possible abuse of a principle does not negate its necessity. If sovereignty is absolute, then no internal genocide, no systematic persecution can legitimise external action. If, on the other hand, fundamental rights are truly universal, they must prevail when a state becomes an extermination machine.

The crux of the matter is all here: either we recognise that sovereignty is conditioned by a minimum and unbreakable respect for human rights, or we admit that international law is not a right of the individual, but merely a pact between governments. In the first case, it is necessary to clearly reformulate the criteria that make intervention to protect threatened populations legitimate, removing them as far as possible from political arbitrariness and the game of vetoes. In the second case, we must have the courage to say that international law, as conceived, primarily protects states, not human beings.

It is not a matter of wishing for a world in which war becomes an ordinary instrument of moralisation. It is about unravelling a contradiction. An order that prohibits aiding those who are massacred because their executioner acts ‘within its borders’ cannot be said to be completely just. The border is not a sacrament. It is a historical fact, changeable, sometimes arbitrary. Human life, on the other hand, is the very foundation of every right.

If international law wants to remain true to its universal ambition, it must emancipate itself from this ambiguity. It must recognise that sovereignty is not an unconditional privilege, but a responsibility. When it turns into systematic persecution, it loses its legitimacy. Otherwise, we will continue to invoke great principles in diplomatic halls, while outside, in silence, massacres are being carried out that the law, out of an excess of respect for borders, has chosen not to see.

And so the choice is clear: either we radically reformulate international law, placing the person at the centre and not just the state, or we accept that it remains a balance between powers, useful in preventing certain wars but powerless – and sometimes complicit – in the face of the exterminations committed within domestic walls. A law that does not know how to protect the innocent from organised violence risks becoming, despite itself, the most refined justification for arbitrariness.

Sovereignty cannot be an unlimited right, because it is not an end, but a means. It was created to guarantee order, security, freedom; if a state is permanently transformed into the opposite of all this, i.e. into an apparatus of systematic persecution, then that means betrays the purpose for which it was recognised. At that point, sovereignty does not ‘disappear’ as a political fact, but loses its claim to be an absolute legal shield.

Hence the first reformulation: to make it explicit, at the heart of international law, that there is a threshold of crimes beyond which non-interference is no longer a superior but a subordinate principle. It is not enough to say, in a generic way, that ‘human rights are universal’: it is necessary for the legal system to have a lock, and therefore also a key. The lock is the strict definition of what triggers the protection of sovereignty; the key is a credible procedure, not left to the chance of alliances.

The threshold, in essence, must concern not injustice as such – because injustice is everywhere, and no world would be more at war than the one in which all injustice authorises the use of force – but injustice that becomes a system: exterminations, deportations, ethnic cleansing, organised mass repression, generalised persecution, the deliberate destruction of the minimum living conditions of a section of the population. When a political power enters this zone, we are no longer in the field of ‘violations’ as correctable defects, but in the field of the transformation of the state into an instrument of annihilation.

However, even if the threshold were well defined, the crux of the decision remains. As long as the decision depends on a body in which a few actors can block everything out of national interest, the norm remains noble and impotent. And here we need institutional courage: either we accept that the Security Council is a ‘political’, and therefore partial, arbiter, or we build, alongside or within the UN, a mechanism that makes inaction more difficult when mass crimes are at stake. There is no need for a world government; a simple and strict procedural principle would suffice: on proven mass crimes, the power of veto cannot count as a right of total paralysis. If politics wants to stop action, it must be able to do so in the light of day, with controllable reasons; or it must publicly choose not to do so, bearing the moral and diplomatic cost of inaction.

But such a reform alone would generate another risk: turning intervention into a military automatism. Then the second leg of such a hypothetical reform is needed, that of guarantees. International law should legitimise external action not as punishment, not as crusade, but as limited and purposeful protection. This clearly means that the use of force should be a last resort, and always proportionate to the purpose of interrupting the extermination machine, opening humanitarian corridors, securing populations and creating minimum conditions for a political transition. In extreme and unendurable cases, even that of overthrowing a criminal regime. If the objective becomes ‘remaking the country’ in the image of the intervener, one falls back into imperial vice; but if the objective remains immediate and verifiable protection, force becomes an exceptional tool, not a project of domination.

In this perspective, the legitimacy of the intervention is not only measured in the moment of the decision, but in the subsequent conduct. And this is where international law should be ruthlessly consistent: a declared ‘humanitarian’ intervention that deliberately produces indiscriminate suffering, that uses methods incompatible with the protection of civilians, that remains unmonitored and unaccountable, must lose its claim to morality. In other words, the reform does not consist in granting a broader licence to the use of force, but in tying it to tighter, controllable, ex-post judgable constraints.

Finally, there remains a point that is often pretended to be ignored: intervention is not only military. It is also prevention, diplomatic pressure, international justice, support for civil societies, protection of refugees, interdiction of arms flows that feed the massacres. Serious international law should make internal oppression ‘costly’ even before it reaches breaking point. Because the real scandal is not just the absence of final action; it is the habit of daily tolerance, which lasts for years, until tragedy becomes routine.

Such a reform would not be an idealistic whim, but an assumption of responsibility, the fruit of an awareness of a need for consistency, because a law that recognises the universality of man and then stops at the executioner’s doorstep is not universal: it is merely prudish, opportunistic and essentially devious. And such a law is sooner or later perceived not as law, but as a mask, and in particular as a valuable mask for any enemy of a true international law, and a particularly dangerous enemy, because it aims to erode its foundations and destroy it from within.